13

Journal of the association of physicians of india may 2014 VOL. 62

Original Article

A Clinical Trial of Nitrosense Patch for the

Treatment of Patients with Painful Diabetic
Neuropathy
RP Agrawal1, S Jain2, S Goyal3, S Singhal4, L Lindgren5, E Sthengel6
Abstract
Aims: Impaired nitric oxide synthesis has been implicated as one of the underlying causes of diabetic
painful neuropathy (DPN). Hence, effects of a cutaneous, nitric oxide releasing patch (NitroSense Derma
Protect) were evaluated in subjects with DPN.
Methods: Fifty diabetics were randomised to active/placebo arms after a 2 wk wash-out period. Patients
received 24 mg patches (each patch releases around 9 nmol/cm 2/min of nitric oxide) for 3 hrs, every
other day during a 3 wks period, or indistinguishable placebo patches. The extent of pain was recorded at
start, at each visit and following completion of the study. Changes in pain from baseline were measured
using the 11 point lickert scale (pls), visual analogue scale (vas), short form mcgill pain questionnaire
(sf-mpq), present pain intensity (PPI) scale.
Results: Subjects treated with patch experienced a statistically significant reduction in pain from baseline
when compared to placebo (PLS scale; p = 0.05). Defining responders as subjects with a > 50% reduction
in PLS score from baseline, the number needed to treat (NNT) was calculated as 3.0. A significant posttreatment decrease (p = 0.009) in vibration perception threshold (VPT) for left foot after active treatment
was observed.
Conclusions: Present results highlight utility of NitroSense Derma Protect as controllable nitric oxide
source for patients with DPN.

Introduction

iabetic painful neuropathy (DPN) is a major complication of both type 1 and type
2 diabetes mellitus. During the course of the condition, 20 to 90% of patients
develop diabetic neuropathy 1 and 7.5% of diabetes patients present with neuropathy
when diagnosed; this figure rises to 45% after 25 years. 2 In type 1 diabetes mellitus
pain typically becomes symptomatic after years of chronic hyperglycaemia, while pain
in patients with type 2 diabetes mellitus manifests within a few years of diagnosis.
Professor, 2Research Scientist,
Research Fellow, 4Statistician,
Diabetes Care and Research
Centre, S.P. Medical College,
Bikaner, Rajasthan, India;
5
Director Research and Dev.,
6
Laboratory Manager, NOLabs
AB, Rnnowsgatan 8, S-252 25
Helsingborg, Sweden
Received; 29.08.2012;
Revised: 12.02.2013;
Accepted: 02.07.2013
1

JAPI MAY 2014 VOL. 62

DPN has a variable presentation with some patients experiencing painful neuropathy
while others have asymptomatic, progressive loss of peripheral nerve function. 3 Pain in
these patients manifests in various forms including burning, sharpness, tingling or deep
lancinating pain and may develop into severe, unremitting pain with exacerbations at
night. In more severe cases, patients also suffer from sleep deprivation and depression. 4
Several studies have implicated impaired nitric oxide synthesis, i.e. disturbance
of endogenous nitric oxide production, as a contributing factor to the pathogenesis
of DPN. In rats, impaired neuronal nitric oxide generation induced hyperalgesia, i.e.
extreme sensitivity to pain. 5,6 There is also evidence that impaired blood flow plays a
role and clinical data from type 2 diabetes patients with DPN has shown that decreased
nitric oxide production influences endoneural blood flow. 7 The importance of nitric

385

14

oxide in neuropathic pain is further substantiated

by recent studies using various nitrate formulations,
including, isosorbide dinitrate (ISDN) and glyceryl
dinitrate (GTN). 8,9
NitroSense derma protect is a nitric oxide eluting
patch which was registered as a Class I medical
device for use as protective coverage to prevent the
occurrence of ulcers on thinned skin in patients with
diabetes or other diseases with skin involvement.
Accordingly, NitroSense derma protect is applied on
intact skin to directly cover and protect skin under the
device. Pure nitric oxide is slowly released from the
device, over several hours. In addition to its physical
protective effects on skin, NitroSense derma protect
is also protected from microbial invasion by nitric
oxide. 10-12 A further effect of applying NitroSense
derma protect to skin is the by nitric oxide mediated
local vasodilatation that results in increased blood
flow. In the proximity of hypoxic nerve endings found
in the DPN patients, the nitric oxide mediated increase
in blood flow might result in relief of neuropathic
pain. The extent to which nitric oxide released from
NitroSense derma protect directly applied to a
diabetic foot exerts any effects has not previously
been investigated.
The primary objective of this study was to evaluate
the feasibility of repeated NitroSense derma protect
patch usage to protect skin on feet or at other locations
in patients with diabetic neuropathy, including
an assessment of the products safety. In addition,
and considering the role of nitric oxide in the
pathophysiology of neuropathy, the potential effects
of nitric oxide on neuropathic pain relief were also
studied as a secondary objective.

Subjects
The study design was based on prior experience
gained from similar studies using other nitric oxide
releasing substances. 13-17 Results from these studies
suggested that patients need at least three consecutive
weeks of nitric oxide treatment in order to experience
initial signs of significant decrease in neuropathic
pain.
Selection of Study Population

The present study was a double-blind, randomised,

placebo controlled study enrolling 50 type 2 diabetes
mellitus patients diagnosed with DPN (Clinical
Investigation No. EX-260808-01). The trial complied
with the Declaration of Helsinki and was approved by
the Ethics Committee. Informed consent was obtained
from all study participants.
Inclusion Criteria

a. Male or female
b. Age 18

386

Journal of the association of physicians of india may 2014 VOL. 62

c. Stable glycaemic control

d. No other medication for neuropathy for at least
two weeks before inclusion
e. At l e a s t t h r e e m o n t h s o f n e u r o p a t h i c p a i n
symptom duration
f. Peripheral neuropathy involving only lower limbs
Exclusion Criteria

a. Erratic glycaemic control

b. Peripheral vascular disease with absent foot
pulses
c. Presence of foot ulceration
d. Treatment with sub lingual glyceryl trinitrate
e. Male patient on sildenafil therapy or any other
vasoactive medicine
f. Presence of other causes of neuropathy like
alcohol, renal impairment, drugs and
toxins, nutritional deficiency
g. Presence of other causes of pain
h. Presence of skin disease, such as atopic dermatitis,
psoriasis
i.

Symptoms/diagnosis of depression

j. Patients with cardiac ischaemic and postural

hypotension.
Prior and Concomitant Therapy

Patients underwent a two week wash-out period

if they were taking any prior medication for the
treatment of neuropathic pain. Subjects were allowed
to continue current drug therapy to manage diabetes.
Use of pain medication during the study period was
avoided, however, in case where rescue medication
was needed, NSAIDs or paracetamol-based drugs
were to be used.
At the start, baseline information from anamnesis
(medical history), neurological (including pain) and
physical examinations, as well as from appropriate
laboratory and physiological tests * were collected
for each patient. Peripheral vascular disease was
excluded based on clinical history and on patients
ankle brachial index (ABI). Patients with adverse
events including excessive skin reactions or allergic
responses were removed from the study and not
included in the assessment, as were the non-compliant
patients.

Material and Method

Treatment Administered

Patient treatments were performed at the Diabetes

Care and Research Centre, SP Medical College,
Bikaner, India and application time of NitroSense
derma protect was controlled by a nurse administering
* Blood glucose, blood urea, serum creatinine, HbA1c, lipid profile, urine
for microalbuminuria, fundus, ECG and chest X-ray.

JAPI MAY 2014 VOL. 62

15

Journal of the association of physicians of india may 2014 VOL. 62

Total Patients
n=50

Two weeks
washout
period

Neurological examination
Biochemical Investigations
Pain assessment scores (SFMPQ, VAS, PPI and Lickert Scale)
Nerve conduction velocity studies

Randomly selected
Group A (Study group)
n=25
NitroSense Derma Protect
dressing (every second day /
week for 3 weeks) on each
foot dorsun

Group B (Placebo Group)

n=25
Placebo dressing (every
second day / week for 3
weeks) on each foot dorsum

Withdrawn due to skin

reaction (n=1)

Withdrawn due to non

compliance (n=1)

Completed trial (n=24)

Completed trial (n=24)

After 3 weeks
Neurological examination
Biochemical Investigations
Pain assessment scores (SFMPQ, VAS, PPI and Lickert Scale)
Nerve conduction velocity studies

treatment groups, and generated the randomisation

list. The latter was done using dedicated computer
software. In order to assure blinding of the study,
special care was given to use a placebo patch with
identical appearance to the active NitroSense derma
protect patch.
Dose Selection

Each NitroSense derma protect patch contains 24

mg of a nitric oxide donor. In-vitro measurements of
release rates of the 24 mg NitroSense derma protect
patch showed a release of 0.14 mole of nitric oxide
per minute or 9 nmole/cm 2 of nitric oxide per minute.
Efficacy and safety assessments

To assess the safety of NitroSense derma protect,

skin changes were evaluated by physical examination
at the start, at pre-determined time points during
the study, and at the end of the study. The efficacy
of topical nitric oxide administered by NitroSense
Derma Protect applied to skin was assessed on (1)
neuropathic pain and (2) on the pathophysiology
of diabetic neuropathy. Pain relief was measured
using validated pain score methods, including visual
analogue scale (vas ), present pain intensity (ppi ),
11 point lickert scale (p l s ), and the short form
McGill pain questionnaire (SF-MPQ). Neurological
examinations and nerve conduction studies were
performed on the lower limb in all subjects at the
start of the study and at follow-up after three weeks
of usage of NitroSense derma protect. Neurological
examination consisted of testing and grading of deep
tendon reflexes, checking for sensory neuropathy
using a Semmes-Weinstein monofilament 5.07 (10 gm)
and vibration testing using a biothesiometer.

Statistical Methods
Statistical Analysis (NNT; paired and unpaired students t test)

Fig. 1 : Flow chart

and removing the patch. Prior to each application,

NitroSense derma protect patches were prepared by a
trained nurse or another trained medical professional
according to instructions provided in the protocol.
Treatment was initiated by applying NitroSense
derma protect to the dorsum of the each foot or
another painful area on each limb of a subject. Patches
were left in situ for three hours there after NitroSense
derma protect was removed by a nurse. Treatment
was repeated every second week-day (i.e. Mondays,
Wednesdays and Fridays), during three consecutive
weeks (Figure 1).
Methods of assigning patients to treatment group

An independent statistician provided the forcing

block size, defined the patient distribution between

JAPI MAY 2014 VOL. 62

Paired and un-paired students t-test were used

for the statistical analysis of pain score (PLS, VAS,
SFMPQ and PPI) and vibration perception threshold
(VPT) data obtained from 48 subjects who completed
the study. Statistical analyses were performed on
different data sets comparing pain scores or VPT
values at baseline (before treatment) to pain scores or
VPT values after ten treatments (after three weeks).

Results
A total of 50 subjects were enrolled in the study
and 48 (96%) completed the three weeks of the study.
After un-blinding there were 24 patients in each of
active and placebo groups. One of the two subjects
who did not complete the study was withdrawn due
to a skin reaction; the other was excluded due to noncompliance. Except for the above mentioned subjects,
no other complaints such as irritation or itching were
reported. None of the subjects requested to change

387

16

Journal of the association of physicians of india may 2014 VOL. 62

SFMPQ

PLS
0,5

1,0

0,0

0,0

-0,5

-1,0
-2,0

-1,0

-3,0

-1,5

-4,0

-2,0

-5,0
0

10

VAS

10

10

PPI

0,5

0,5

0,0
0,0

-0,5
-1,0

-0,5

-1,5

-1,0

-2,0
-2,5
0

10

-1,5
0

Fig. 2 : Comparisons of changes in pain scores over time (all subjects) each graph represents data from 48 patients assessed
by one of the four different pain scales. Changes in pain score from baseline (y axis) are plotted vs. time (number of
visits, x-axis). Filled squares = placebo, Un-filled rhomboids = active
Table 1 : Demographic and other baseline characteristics
Parameter
Number of patients
Age, median (range)
Weight (Kg), mean (range)
Height (cm), mean (range)
BMI, mean (range)
Ankle brachial index
HbA1c (%)
Fasting blood sugar (mg/dl)
Dyslipidaemia (mg/dl)

Active
24
52 (26 -75)
63.9 (44 88)
160,0 (144 176)
25.0 (17,2 32,9)
R 1.21 (1.07-1.35), L 1.16 (0.62-1.37)
9.1 (5.6-13.4)
234.5 (93.2-442.6)
211.5 (148.5-344.2)

Placebo
24
55 (39 75)
66.1 (45 79)
161,3 (146 175)
25.9 (17,1 35,7)
R 1.21 (0.96-1.38), L 1.18 (0.92-1.33)
9.67 (5.6-13.0)
239.36 (106.0-380.7)
197.93 (146.5-289.5)

the place of application, however, skin discolouration

was observed in some cases. Despite no complaint by
subjects, the place for application were changed and
discolouration disappeared in 4-5 days.

There was a significant post-treatment decrease

(from 27.50 2.61 to 22.21 2.21; p = 0.009) in VPT
for the left foot following treatment with NitroSense
derma protect.

All subjects completing three weeks of NitroSense

derma protect treatment experienced a reduction
in pain (Table 2). Moreover, pain reduction was
statistically significant when measured using the PLS
scale, comparing active and placebo responses (p =
0.05, Table 2, Figure 2). Pain reduction as assessed by
the VAS, SF-MPQ or PPI scales was not statistically
significant. A subset analysis that excluded patients
with mild pain (PLS < 4) showed a larger difference
between active and placebo but the difference was still
only significant for the PLS score, most likely due to
a too low number of evaluated subjects (n = 17 active
and 19 placebo, Table 3).

Analysis of the nerve conduction velocity did not

show any significant difference among all groups
analysed.

388

The number needed to treat (NNT) was 3.0 when

calculated using the change in PLS scores from
baseline and defining responders as > 50% pain relief
(Table 4).
Only one patient receiving NitroSense derma
protect experienced skin reaction which subsided
spontaneously without any intervention.

JAPI MAY 2014 VOL. 62

17

Journal of the association of physicians of india may 2014 VOL. 62

Table 2 : Comparison of pre- and post treatment pain score in the active and placebo arm
Pre-treatment

Evaluation tool
ACTIVE (n=24)
PLS
VAS
SFMPQ
PPI
PLACEBO (n=24)
PLS
VAS
SFMPQ
PPI

PLS
VAS
SFMPQ
PPI

Improvement from baseline

se

Mean

se

Diff

p*

5.08
5.38
31.04
2.79

0.41
0.42
3.26
0.21

3.42
3.63
27.29
1.71

0.33
0.28
3.03
0.19

-1.67
-1.75
-3.75
-1.08

7.78
5.04
3.10
6.84

0.001
0.001
0.0051
0.001

0.35
3.88
0.27
0.39
3.58
0.32
3.26
22.67
2.38
0.14
1.96
0.14
Comparison of Active and Placebo response
Active (n=24)
Placebo (n=24)
Mean Change
Mean Change
se
se
from baseline
from baseline
-1.67
0.21
-1.13
0.16
-1.75
0.35
-1.42
0.22

-1.13
-1.42
-2.21
-0.87

6.91
6.55
2.20
7.00

0.001
0.001
0.0384
0.001

5.00
5.00
24.88
2.83

Evaluation tool

**

Post-treatment

Mean

-3.75
-1.08

1.21
0.16

-2.21
-0.88

1.01
0.13

Active placebo comparison

Diff

p**

0.54
0.33

2.01
0.82

0.05
0.42

1.54
0.20

0.98
1.03

0.33
0.31

Unpaired, two-tailed t-test; *Paired, two-tailed t-test

Discussion
Diabetic neuropathy is a symmetrical peripheral
polyneuropathy that results from nerve damage after
prolonged periods of suboptimal glycaemic control. 18
Diabetic neuropathy has been defined as presence of
symptom and signs of peripheral nerve dysfunction
in diabetes after exclusion of other causes, including
hereditary, traumatic, compressive, metabolic,
toxic, nutritional, infectious, immune mediated and
neoplastic causes, as well as, causes secondary to other
systemic illness. 19
In the absence of a curative therapy, treatment efforts
are directed towards providing symptomatic pain
control using pharmacological tools. Pharmacological
therapy includes serotoninnorepinephrine reuptake
inhibitors (duloxetine and velafaxine), tricyclic
antidepressants (amitryptiline), narcotic analgesic
(oxycodone CR), anticonvulsants (pregabalin,
Gabapentin, carbamazepine and lamotrigine) and
topical agents (capsaicin and lidocaine) but none of
them is entirely satisfactory. 20,21 This present study
is comparable to another study conducted using
amitryptiline in which NNT was found to be 4.1. 22
Our previous double-blind, cross-over trial has
demonstrated the utility of a GTN spray, for the
treatment of pain due to DPN. In this study, 43
subjects with pain due to DPN were divided in to two
cohorts. Cohort A received GTN for four weeks and
cohort B received placebo. After two weeks, subjects
in each cohort were crossed over and treatment
continued for another four weeks. Overall, subjects
in both cohorts experienced significant (p < 0.001)

JAPI MAY 2014 VOL. 62

improvements in their pain scores while receiving

GTN when compared to subjects receiving placebo. In
addition, after crossing over into the GTN arm from
placebo, subjects observed significant improvements
(p < 0.001) in all pain scores. The NNT (using VAS
scores) in this study was calculated as 4. The drug was
well tolerated by all the patients except palpitation
and headache for some days in five patients. 16
More recently we completed a study regarding
clinical utility of topically applied nitrates for the
treatment of DPN. For example, we have evaluated
the safety and efficacy of sodium valproate and GTN
alone, as well, as in combination in patients with
DPN. In this study, 80 subjects were divided in to
four cohorts, each receiving either GTN, sodium
valproate, both or placebo. After three months,
subjects in all three cohorts experienced significant (p
< 0.001 to p < 0.05) improvements in pain scores along
with improvements in certain electrophysiological
parameters. The NNT (using VAS scores) for sodium
valproate, GTN and their combination in this study
were calculated as 7, 5 and 4, respectively. These
results suggest combining drugs with different
mechanisms of action may achieve synergistic
analgesic effects in DPN. 17
Others have also observed antinociceptive effects
of topically applied nitric oxide donors in patient
with DPN. For example, glyceryl trinitrate (GTN),
a nitric oxide donor with local vasodilatory effects,
is bio-transformed to release nitric oxide that can
activate guanylyl cyclase and increases the synthesis
of cyclic-GMP in smooth muscle and other tissues. The
pharmacological and biochemical effects of glyceryl

389

18

Journal of the association of physicians of india may 2014 VOL. 62

Table 3 : Comparison between pain score improvement in active and placebo treatment groups. Only patients with severemoderate pain included in the comparison
Evaluation tool
PLS
VAS
SFMPQ
PPI

Active (n=17)
Mean change
se
from baseline
-1.94
0.23
-2.35
0.33
-4.53
1.27
-1.24
0.17

Placebo (n=19)
Mean change
se
from baseline
-1.26
0.16
-1.47
0.23
-2.68
2.55
-0.89
0.13

Active placebo comparison

Diff

p*

0.68
0.88
1.85
0.35

2.07
1.91
1.00
1.37

0.05
0.065
0.33
0.18

* un-paired, two-tailed t-test

Table 4 : % responders and NNT calculated for the active and placebo groups using the change in PLS score from baseline
Whole group (n=48)
Responders
Active
Placebo
62 %
29 %
29 %
0%

Criteria for positive outcome

30% reduction from baseline
50% reduction from baseline

trinitrate spray appear to be identical to those of

endothelial-derived relaxing factor. 23
In summary, NitroSense derma protect is a highly
promising device for the treatment of neuropathic
pain in patients with DPN and additional studies
should be conducted. The present study demonstrates
the therapeutic utility of locally delivered nitric oxide.
Furthermore, and in contrast to other topical nitric
oxide formulations, NitroSense derma protect is
safe, well tolerated and does not have any systemic
side effects. An extended treatment with NitroSense
derma protect to eight or more weeks, selecting
patients with moderate to severe pain and/or better
glycaemic control at baseline may result in even more
pronounced pain relief.

Acknowledgement
We are thankful to NO Labs AB, Sweden for
supplying Nitrosense derma patch.

References
1.

Brown MJ, Asbury AK. Diabetes neuropathy. Ann Neurol. 1984;15:2-12.

2.

Pirat J. Diabetes mellitus and its degenerative complications: a study

of 4400 patients observed between 1947 and 1973. Diabetes Care
1978;1:168-186.

3.

Galli V, Firenze C, Mazzotta G, Delgatto S. Neuropathy in children and

adolescence with diabetes mellitus. Acta Neurol Scand. 1988;78:136140.

4.

Watkins PJ. Pain and diabetic neuropathy. Br Med J. 1984;288:168-169.

5.

Sasaki T, Yasuda H, Maeda K, Kikkawa R. Hyperalgesia and decreased

neuronal nitric oxide synthase in diabetic rats. NeuroReport
1998;9:243247.

6.

Rodella L, Rezzani R, Corsetti G and Bianchi R. Nitric oxide

involvement in the trigeminal hyperalgesia in diabetic rats. Brain
Research 2000;865:112-115.

7.

Pitei DL, Watkins PJ, Edmonds ME. NO-dependent smooth muscle

vasodilatation is reduced in NIDDM patients with peripheral sensory
neuropathy. Diabet Med. 1997;14:284-290.

8.

Yuen KC, Baker NR, Rayman G. Treatment of chronic painful diabetic

neuropathy with isosorbide dinitrate spray: a double-blind placebocontrolled cross-over study. Diabetes Care 2002;25:1699-703.

9.

Rayman G, Baker NR, Krishnan ST. Glyceryl trinitrate patches as an

alternative to isosorbide dinitrate spray in the treatment of chronic

390

Mod-severe group (n=36)

Responders
NNT
Active
Placebo
3.4
47 %
21 %
3.8
3.0
24 %
0%
4.3
painful diabetic neuropathy. Diabetes Care 2003;26:2697-8.

NNT

10. Ghaffari A, Neil DH, Ardakani A, Road J, Ghahary A, Miller CC. A direct
nitric oxide gas delivery system for bacterial and mammalian cell
cultures. Nitric Oxide 2005;12:129-40.
11. Ghaffari A, Miller CC, McMullin B, Ghahary A. Potential application
of gaseous nitric oxide as a topical antimicrobial agent. Nitric Oxide
2006;14:21-9.
12. Nablo BJ, Schoenfisch MH. Poly(vinyl chloride)-coated sol-gels for
studying the effects of nitric oxide release on bacterial adhesion.
Biomacromolecules 2004;5:2034-41.
13. Kochar DK, Agrawal RP, Joshi A, Kumawat BL. Herpes Zoster and
Post Herpatic Neuralgia A clinical trial of aspirin in chloroform for
anodyne. JAPI 1998;46:337-340.
14. Kochar DK, Jain N, Agrawal RP, Srivastava T, Agarwal P, Gupta S.
Sodium valproate in the management of painful neuropathy in type
2 diabetes a randomized placebo controlled study. Acta Neurol
Scand. 2002;106:248-252.
15. Kochar DK, Rawat N, Agrawal RP, Vyas A, Beniwal R. Sodium valproate
for painful diabetic neuropathy: a randomized double-blind placebocontrolled study. Q J M 2004;97:33-38.
16. Agrawal RP, Choudhary R, Sharma P, Sharma S, Beniwal R, Kaswan K,
Kochar D.K. Glyceryl trinitrate spray in the management of painful
diabetic neuropathy: A randomized double blind placebo controlled
cross-over study. Diabetes Research and Clinical Practice 2007;77:161167.
17. Agrawal RP, Jitender Goswami, Shreyans Jain, Kochar DK.
Management of diabetic neuropathy by sodium valproate and
glyceryl trinitrate spray: A prospective double-blind randomized
placebo-controlled study. Diabetes Research and Clinical Practice
2009;83:371-378.
18. Hoke A, Feasby T. Disorders of the peripheral nervous system, in:
H. Homes (Ed.), Kelleys Textbook of Internal Medicine, Lippincolt
Williams and Wilkins. Philadelphia 2000;29802981.
19. Bulton AJM, Malik RA. Diabetic neuropathy. Med Clin North Am
1998;82.
20. Low PA, Dosto RR. Symptomatic treatment of painful neuropathy
(editorial). JAMA 1998;280:1863.
21. Jensen PG, Larsen JR. Management of painful diabetic neuropathy.
Drugs 2001;18:739749.
22. Moore RA, Derry S, Aldington D, Cole P, Wiffen PJ, Amitriptyline for
neuropathic pain and fibromyalgia in adults. Cochrane Database
Syst Rev. 2012.
23. Molina C, Anderson JW, Rapaport RM, Waldman SA, Murad F. Effects
of in vivo nitroglycerin therapy on endotheliumdependent and
independent relaxation and cyclic GMP accumulation in rat arota. J
Cardiovasc Pharmacol 1987;10:371-378.
JAPI MAY 2014 VOL. 62